{"id":"https://openalex.org/W2938171278","doi":"https://doi.org/10.1145/3300053.3319419","title":"Characterizing CUDA Unified Memory (UM)-Aware MPI Designs on Modern GPU Architectures","display_name":"Characterizing CUDA Unified Memory (UM)-Aware MPI Designs on Modern GPU Architectures","publication_year":2019,"publication_date":"2019-04-10","ids":{"openalex":"https://openalex.org/W2938171278","doi":"https://doi.org/10.1145/3300053.3319419","mag":"2938171278"},"language":"en","primary_location":{"id":"doi:10.1145/3300053.3319419","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3300053.3319419","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 12th Workshop on General Purpose Processing Using GPUs","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5061582464","display_name":"Karthik Vadambacheri Manian","orcid":"https://orcid.org/0000-0001-7937-7035"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"K. V. Manian","raw_affiliation_strings":["The Ohio State University Columbus, Ohio"],"affiliations":[{"raw_affiliation_string":"The Ohio State University Columbus, Ohio","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027838642","display_name":"Anis Ammar","orcid":"https://orcid.org/0000-0002-6588-8739"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"A. A. Ammar","raw_affiliation_strings":["The Ohio State University Columbus, Ohio"],"affiliations":[{"raw_affiliation_string":"The Ohio State University Columbus, Ohio","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012040774","display_name":"Amit Ruhela","orcid":"https://orcid.org/0000-0001-6547-714X"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"A. Ruhela","raw_affiliation_strings":["The Ohio State University Columbus, Ohio"],"affiliations":[{"raw_affiliation_string":"The Ohio State University Columbus, Ohio","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5056955016","display_name":"C-H. Chu","orcid":null},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"C.-H. Chu","raw_affiliation_strings":["The Ohio State University Columbus, Ohio"],"affiliations":[{"raw_affiliation_string":"The Ohio State University Columbus, Ohio","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034293705","display_name":"Hari Subramoni","orcid":"https://orcid.org/0000-0002-1200-2754"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"H. Subramoni","raw_affiliation_strings":["The Ohio State University Columbus, Ohio"],"affiliations":[{"raw_affiliation_string":"The Ohio State University Columbus, Ohio","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5024879682","display_name":"Dhabaleswar K. Panda","orcid":"https://orcid.org/0000-0002-0356-1781"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"D. K. Panda","raw_affiliation_strings":["The Ohio State University Columbus, Ohio"],"affiliations":[{"raw_affiliation_string":"The Ohio State University Columbus, Ohio","institution_ids":["https://openalex.org/I52357470"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5061582464"],"corresponding_institution_ids":["https://openalex.org/I52357470"],"apc_list":null,"apc_paid":null,"fwci":2.8892,"has_fulltext":false,"cited_by_count":15,"citation_normalized_percentile":{"value":0.90699191,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":91,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"43","last_page":"52"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9998999834060669,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9998999834060669,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11181","display_name":"Advanced Data Storage Technologies","score":0.9991000294685364,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12808","display_name":"Ferroelectric and Negative Capacitance Devices","score":0.9961000084877014,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.8876543045043945},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8750574588775635},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6689715385437012},{"id":"https://openalex.org/keywords/programmer","display_name":"Programmer","score":0.6197881698608398},{"id":"https://openalex.org/keywords/pascal","display_name":"Pascal (unit)","score":0.49142396450042725},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.48033061623573303},{"id":"https://openalex.org/keywords/supercomputer","display_name":"Supercomputer","score":0.45640408992767334},{"id":"https://openalex.org/keywords/programming-paradigm","display_name":"Programming paradigm","score":0.4405975937843323},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.2589821219444275},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.11731529235839844}],"concepts":[{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.8876543045043945},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8750574588775635},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6689715385437012},{"id":"https://openalex.org/C2778514511","wikidata":"https://www.wikidata.org/wiki/Q1374194","display_name":"Programmer","level":2,"score":0.6197881698608398},{"id":"https://openalex.org/C75608658","wikidata":"https://www.wikidata.org/wiki/Q44395","display_name":"Pascal (unit)","level":2,"score":0.49142396450042725},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.48033061623573303},{"id":"https://openalex.org/C83283714","wikidata":"https://www.wikidata.org/wiki/Q121117","display_name":"Supercomputer","level":2,"score":0.45640408992767334},{"id":"https://openalex.org/C34165917","wikidata":"https://www.wikidata.org/wiki/Q188267","display_name":"Programming paradigm","level":2,"score":0.4405975937843323},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.2589821219444275},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.11731529235839844}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3300053.3319419","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3300053.3319419","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 12th Workshop on General Purpose Processing Using GPUs","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":28,"referenced_works":["https://openalex.org/W1541729506","https://openalex.org/W1825216778","https://openalex.org/W1964981582","https://openalex.org/W2024565525","https://openalex.org/W2025388037","https://openalex.org/W2041679393","https://openalex.org/W2078391824","https://openalex.org/W2322990374","https://openalex.org/W2402549220","https://openalex.org/W2409018029","https://openalex.org/W2559253174","https://openalex.org/W2566915031","https://openalex.org/W2580688187","https://openalex.org/W2584650020","https://openalex.org/W2612605147","https://openalex.org/W2624632852","https://openalex.org/W2626009569","https://openalex.org/W2704425079","https://openalex.org/W2740001873","https://openalex.org/W2767028519","https://openalex.org/W2770732609","https://openalex.org/W2889456810","https://openalex.org/W2901994046","https://openalex.org/W2913640523","https://openalex.org/W4206486352","https://openalex.org/W4248595399","https://openalex.org/W4389317874","https://openalex.org/W4403534683"],"related_works":["https://openalex.org/W2379153735","https://openalex.org/W3062287","https://openalex.org/W2384867379","https://openalex.org/W2380390332","https://openalex.org/W2046172023","https://openalex.org/W2972896947","https://openalex.org/W2742145873","https://openalex.org/W1537323515","https://openalex.org/W2047375041","https://openalex.org/W4236300446"],"abstract_inverted_index":{"The":[0],"CUDA":[1,23,37,95],"Unified":[2],"Memory":[3],"(UM)":[4],"interface":[5],"enables":[6],"a":[7,113],"significantly":[8],"simpler":[9],"programming":[10,35],"paradigm":[11],"and":[12,45,97,146,154,171,195,204,222,227],"has":[13,74,106],"the":[14,19,26,40,54,83,101,117,130,148,188,201,234],"potential":[15],"to":[16,31,79,115,127,155,163,247],"fundamentally":[17],"change":[18],"way":[20],"programmers":[21],"write":[22],"applications":[24,78],"in":[25,34,53,60,66,122,133],"future.":[27],"Although":[28],"UM":[29,52,88,105,121,184,212],"leads":[30],"high":[32,84,91],"productivity":[33,85],"using":[36],"by":[38,87,104,225],"simplifying":[39],"programmer's":[41],"view":[42],"of":[43,57,82,120,124,150,176,210],"CPU":[44,172],"GPU":[46,98,170,177,190],"memory":[47],"spaces,":[48],"initial":[49],"support":[50],"for":[51,173,230,237],"Kepler":[55,189],"series":[56],"GPUs":[58],"lacked":[59],"performance":[61,102,118,149,198,208],"necessitating":[62],"several":[63],"UM-aware":[64,131,151],"designs":[65,132,185,213,219],"state-of-the-art":[67,134],"MPI":[68,77,135,152,160,224],"runtimes":[69,96,136,161],"such":[70],"as":[71,94],"MVAPICH2-GDR.":[72],"This":[73],"enabled":[75],"end":[76],"take":[80,141],"advantage":[81],"promised":[86],"along":[89],"with":[90,165],"performance.":[92],"However,":[93],"architectures":[99],"advance,":[100],"offered":[103],"also":[107,241],"improved":[108],"significantly.":[109],"Thus,":[110],"there":[111],"is":[112,245],"need":[114,162],"re-evaluate":[116],"characteristics":[119],"light":[123],"these":[125],"changes":[126],"understand":[128],"how":[129,159],"must":[137],"be":[138],"adapted.":[139],"We":[140],"up":[142,246],"this":[143],"broad":[144],"challenge":[145],"characterize":[147],"operations":[153],"gain":[156],"insights":[157],"on":[158,169,200,220],"deal":[164],"UM-based":[166],"data":[167],"residing":[168],"different":[174],"generations":[175],"architectures.":[178],"Our":[179],"characterization":[180],"studies":[181],"show":[182,214,242],"that":[183,215,243],"conceived":[186],"during":[187],"era":[191],"still":[192],"stands":[193],"valid":[194],"provide":[196],"valuable":[197],"improvement":[199],"latest":[202],"Pascal":[203],"Volta":[205],"GPUs.":[206],"Furthermore,":[207],"evaluation":[209],"optimized":[211],"they":[216],"outperform":[217],"naive":[218],"MVAPICH2-GDR":[221,244],"Open":[223],"4.2x":[226],"2.8x":[228],"respectively":[229],"Intel":[231],"systems.":[232],"Additionally,":[233],"DD":[235],"experiments":[236],"pure":[238],"device":[239],"transfers":[240],"12.6x":[248],"better":[249],"than":[250],"OpenMPI":[251],"(w/":[252],"UCX).":[253]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":2},{"year":2021,"cited_by_count":3},{"year":2020,"cited_by_count":4},{"year":2019,"cited_by_count":3}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}